Glycated albumin and fructosamine do not improve accuracy of glycaemic control assessment in patients with conditions reported to affect HbA1c reliability

Aims: HbA(1c) testing in African populations may be limited due to high prevalence of hemoglobinopathies, anaemia, malaria and renal impairment. We aimed to assess the performance of glycated albumin (GA) and fructosamine in comparison to HbA(1c) for determining glycaemic control in Africans living with type 2 diabetes. Methods: We compared the relationship between fructosamine, GA, and HbA(1c) with mean continuous glucose monitoring (CGM) glucose and assessed the impact of sickle cell trait (SCT), anaemia and renal impairment on the relationship between each measure and CGM glucose. Results: The overall association of HbA(1c), GA and fructosamine with CGM glucose was similar (r = 0.88 [95%CI: 0.84, 0.91], 0.84 [0.79, 0.88] and 0.84 [0.79, 0.88]), respectively. For detecting those with mean CGM glucose >8 mmol/L HbA(1c) had similar diagnostic accuracy to GA and fructosamine, even in those with conditions reported to affect HbA(1c) performance (n = 63). We found no evidence that SCT (n = 43/192) altered the relationship between HbA(1c), fructosamine or GA with CGM glucose (p > 0.3 for all). However, individuals with anaemia showed an underestimation of CGM glucose by HbA(1c) and fructosamine compared to those without anaemia (p for interaction <0.005 for both). In contrast, GA with average CGM glucose between those with anaemia and those without were not significantly different. Conclusions: Switching to fructosamine or GA is unlikely to improve the accuracy of laboratory glycaemic monitoring beyond that of HbA(1c) in a population with high prevalence of conditions reported to affect HbA(1c) reliability.